Comparative electrical energy production estimates with deep learning methods

In this study, electricity production from nuclear energy was analyzed for six different countries. We can list these countries as the USA, France, Japan, South Korea, Russia and China. The data sets used in the study were obtained from the Our World in Data resource and are based on long-term data on each country's nuclear energy production processes. The importance of the research is that it provides important information about energy policies and strategies by revealing the role of nuclear energy in electricity production and the production differences between countries. Such a comprehensive analysis can help make critical decisions for the energy sector and contribute to predicting future energy production trends. The novelty of the study lies in making nuclear energy production predictions using both statistical and deep learning models. Among the models used are methods such as Naive, SES, Auto.Arima, Holt-Winters, ETS, Thetaf, NNETAR and MLP. The performance of these models was evaluated with metrics such as RMSE, MAE and MAPE. The analysis results showed that the most appropriate model for each country is different; For example, while the NNETAR model gave the most successful results for the USA, the Thetaf model stood out for France. This shows that the energy production dynamics of each country are different and therefore care should be taken in model selection. The contribution of the study to the literature becomes evident by examining the variety of models used in the analysis and forecasting of nuclear energy production and the performance of these models on different countries. Comparing the methods used in energy production provides valuable information for policy makers and researchers. Additionally, the findings obtained in this study allow important inferences to be made on energy security and sustainability issues. Overall, it is believed that this research can help predict future energy production trends more accurately, making a significant contribution to modeling studies in the energy sector.